{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import tqdm\n",
    "import pickle\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"Models/M1_pc.pkl\" , \"rb\") as file :\n",
    "    model1_pc = pickle.load(file )\n",
    "with open(\"Models/M1_ae.pkl\", \"rb\") as file :\n",
    "    model1_ae= pickle.load(file )\n",
    "    \n",
    "with open(\"Models/M2_pc.pkl\" , \"rb\") as file :\n",
    "    model2_pc = pickle.load(file )\n",
    "with open(\"Models/M2_ae.pkl\", \"rb\") as file :\n",
    "    model2_ae= pickle.load(file )\n",
    "    \n",
    "with open(\"Models/M3_pc.pkl\" , \"rb\") as file :\n",
    "    model3_pc = pickle.load(file )\n",
    "with open(\"Models/M3_ae.pkl\", \"rb\") as file :\n",
    "    model3_ae= pickle.load(file )\n",
    "    \n",
    "with open(\"Models/M4_pc.pkl\" , \"rb\") as file :\n",
    "    model4_pc = pickle.load(file )\n",
    "with open(\"Models/M4_ae.pkl\", \"rb\") as file :\n",
    "    model4_ae= pickle.load(file )\n",
    "    \n",
    "with open(\"Models/M5_pc.pkl\" , \"rb\") as file :\n",
    "    model5_pc = pickle.load(file )\n",
    "with open(\"Models/M5_ae.pkl\", \"rb\") as file :\n",
    "    model5_ae= pickle.load(file )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13927, 8)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th>author_name</th>\n",
       "      <th>para_no</th>\n",
       "      <th>para_text</th>\n",
       "      <th>en_para_text</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>source</th>\n",
       "      <th>vector</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0</td>\n",
       "      <td>Rinkimų programa LLRA-KŠS</td>\n",
       "      <td>Electoral program LLRA-KŠS</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>[-0.01677975431084633, -0.10882312804460526, -...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>1</td>\n",
       "      <td>Lietuvos lenkų rinkimų akcijos – Krikščioniškų...</td>\n",
       "      <td>Lithuanian Polish Election Campaigns - Program...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>[0.12838463485240936, -0.35519272089004517, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>2</td>\n",
       "      <td>2016 m. Seimo rinkimai</td>\n",
       "      <td>2016 Seimas elections</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>[0.2894723415374756, -0.6222676634788513, -0.3...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>3</td>\n",
       "      <td>Mieli Lietuvos žmonės!</td>\n",
       "      <td>Dear Lithuanian people!</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>[-0.21554549038410187, -0.11729791015386581, -...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>4</td>\n",
       "      <td>Mūsų partijos veiklos pagrindas yra krikščioni...</td>\n",
       "      <td>The basis of our party's activities is Christi...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>[-0.054763901978731155, -0.3144015073776245, -...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  author_affiliation author_name  para_no  \\\n",
       "0               LLRA        LLRA        0   \n",
       "1               LLRA        LLRA        1   \n",
       "2               LLRA        LLRA        2   \n",
       "3               LLRA        LLRA        3   \n",
       "4               LLRA        LLRA        4   \n",
       "\n",
       "                                           para_text  \\\n",
       "0                          Rinkimų programa LLRA-KŠS   \n",
       "1  Lietuvos lenkų rinkimų akcijos – Krikščioniškų...   \n",
       "2                             2016 m. Seimo rinkimai   \n",
       "3                             Mieli Lietuvos žmonės!   \n",
       "4  Mūsų partijos veiklos pagrindas yra krikščioni...   \n",
       "\n",
       "                                        en_para_text publish_time  \\\n",
       "0                         Electoral program LLRA-KŠS         2016   \n",
       "1  Lithuanian Polish Election Campaigns - Program...         2016   \n",
       "2                              2016 Seimas elections         2016   \n",
       "3                            Dear Lithuanian people!         2016   \n",
       "4  The basis of our party's activities is Christi...         2016   \n",
       "\n",
       "                     source                                             vector  \n",
       "0  party_election_manifesto  [-0.01677975431084633, -0.10882312804460526, -...  \n",
       "1  party_election_manifesto  [0.12838463485240936, -0.35519272089004517, -0...  \n",
       "2  party_election_manifesto  [0.2894723415374756, -0.6222676634788513, -0.3...  \n",
       "3  party_election_manifesto  [-0.21554549038410187, -0.11729791015386581, -...  \n",
       "4  party_election_manifesto  [-0.054763901978731155, -0.3144015073776245, -...  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_parquet(\"Data/LT_ValidationData.parquet\")\n",
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 13927/13927 [24:12<00:00,  9.59it/s]\n"
     ]
    }
   ],
   "source": [
    "large_prediction = []\n",
    "\n",
    "for item in tqdm( df.to_dict(orient=\"records\") ):\n",
    "\n",
    "    vector = np.array( item[ \"vector\"] ).reshape(1 , 1024 )\n",
    "    \n",
    "    m1_pae = model1_ae.predict( vector )[0]  \n",
    "    m1_ppc = model1_pc.predict( vector )[0]\n",
    "        \n",
    "    m2_pae = model2_ae.predict( vector )[0]  \n",
    "    m2_ppc = model2_pc.predict( vector )[0]\n",
    "        \n",
    "    m3_pae = model3_ae.predict( vector )[0]  \n",
    "    m3_ppc = model3_pc.predict( vector )[0]\n",
    "  \n",
    "    m4_pae = model4_ae.predict( vector )[0]  \n",
    "    m4_ppc = model4_pc.predict( vector )[0]\n",
    "        \n",
    "    m5_pae = model5_ae.predict( vector )[0]  \n",
    "    m5_ppc = model5_pc.predict( vector )[0]\n",
    "    \n",
    "    pc_counter = m1_ppc + m2_ppc + m3_ppc + m4_ppc + m5_ppc\n",
    "    row_pc = int( pc_counter >= 2 )\n",
    "    \n",
    "    ae_counter = m1_pae + m2_pae + m3_pae + m4_pae + m5_pae\n",
    "    row_ae = int( ae_counter >= 2 )\n",
    "        \n",
    "\n",
    "    d = {\"PC\" : row_pc , \n",
    "         \"AE\" : row_ae ,\n",
    "         \"LogReg_ae\" : m1_pae , \n",
    "         \"LogReg_pc\" : m1_ppc , \n",
    "         \"NaiveBaies_ae\" : m2_pae , \n",
    "         \"NaiveBaies_pc\" : m2_ppc , \n",
    "         \"SVM_ae\" : m3_pae , \n",
    "         \"SVM_pc\" : m3_ppc , \n",
    "         \"MLP_ae\" : m4_pae , \n",
    "         \"MLP_pc\" : m4_ppc , \n",
    "         \"KNN_ae\" : m5_pae ,\n",
    "         \"KNN_pc\" : m5_ppc\n",
    "                }\n",
    "    item.update( d )\n",
    "    item.pop(\"vector\")\n",
    "\n",
    "    large_prediction.append( item )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13927, 19)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th>author_name</th>\n",
       "      <th>para_no</th>\n",
       "      <th>para_text</th>\n",
       "      <th>en_para_text</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>source</th>\n",
       "      <th>PC</th>\n",
       "      <th>AE</th>\n",
       "      <th>LogReg_ae</th>\n",
       "      <th>LogReg_pc</th>\n",
       "      <th>NaiveBaies_ae</th>\n",
       "      <th>NaiveBaies_pc</th>\n",
       "      <th>SVM_ae</th>\n",
       "      <th>SVM_pc</th>\n",
       "      <th>MLP_ae</th>\n",
       "      <th>MLP_pc</th>\n",
       "      <th>KNN_ae</th>\n",
       "      <th>KNN_pc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0</td>\n",
       "      <td>Rinkimų programa LLRA-KŠS</td>\n",
       "      <td>Electoral program LLRA-KŠS</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>1</td>\n",
       "      <td>Lietuvos lenkų rinkimų akcijos – Krikščioniškų...</td>\n",
       "      <td>Lithuanian Polish Election Campaigns - Program...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>2</td>\n",
       "      <td>2016 m. Seimo rinkimai</td>\n",
       "      <td>2016 Seimas elections</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>3</td>\n",
       "      <td>Mieli Lietuvos žmonės!</td>\n",
       "      <td>Dear Lithuanian people!</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>4</td>\n",
       "      <td>Mūsų partijos veiklos pagrindas yra krikščioni...</td>\n",
       "      <td>The basis of our party's activities is Christi...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  author_affiliation author_name  para_no  \\\n",
       "0               LLRA        LLRA        0   \n",
       "1               LLRA        LLRA        1   \n",
       "2               LLRA        LLRA        2   \n",
       "3               LLRA        LLRA        3   \n",
       "4               LLRA        LLRA        4   \n",
       "\n",
       "                                           para_text  \\\n",
       "0                          Rinkimų programa LLRA-KŠS   \n",
       "1  Lietuvos lenkų rinkimų akcijos – Krikščioniškų...   \n",
       "2                             2016 m. Seimo rinkimai   \n",
       "3                             Mieli Lietuvos žmonės!   \n",
       "4  Mūsų partijos veiklos pagrindas yra krikščioni...   \n",
       "\n",
       "                                        en_para_text publish_time  \\\n",
       "0                         Electoral program LLRA-KŠS         2016   \n",
       "1  Lithuanian Polish Election Campaigns - Program...         2016   \n",
       "2                              2016 Seimas elections         2016   \n",
       "3                            Dear Lithuanian people!         2016   \n",
       "4  The basis of our party's activities is Christi...         2016   \n",
       "\n",
       "                     source  PC  AE  LogReg_ae  LogReg_pc  NaiveBaies_ae  \\\n",
       "0  party_election_manifesto   0   0          0          0              0   \n",
       "1  party_election_manifesto   0   0          0          0              0   \n",
       "2  party_election_manifesto   0   0          1          0              0   \n",
       "3  party_election_manifesto   1   0          0          1              0   \n",
       "4  party_election_manifesto   1   0          0          1              1   \n",
       "\n",
       "   NaiveBaies_pc  SVM_ae  SVM_pc  MLP_ae  MLP_pc  KNN_ae  KNN_pc  \n",
       "0              0       0       0       0       0       0       1  \n",
       "1              0       0       0       0       0       0       0  \n",
       "2              0       0       0       0       0       0       1  \n",
       "3              0       0       1       0       0       0       0  \n",
       "4              1       0       1       0       1       0       0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction_df = pd.DataFrame( large_prediction )\n",
    "print(prediction_df.shape )\n",
    "prediction_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction_df.to_parquet(\"Data/PredictionDF.parquet\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13927, 19)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th>author_name</th>\n",
       "      <th>para_no</th>\n",
       "      <th>para_text</th>\n",
       "      <th>en_para_text</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>source</th>\n",
       "      <th>PC</th>\n",
       "      <th>AE</th>\n",
       "      <th>LogReg_ae</th>\n",
       "      <th>LogReg_pc</th>\n",
       "      <th>NaiveBaies_ae</th>\n",
       "      <th>NaiveBaies_pc</th>\n",
       "      <th>SVM_ae</th>\n",
       "      <th>SVM_pc</th>\n",
       "      <th>MLP_ae</th>\n",
       "      <th>MLP_pc</th>\n",
       "      <th>KNN_ae</th>\n",
       "      <th>KNN_pc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0</td>\n",
       "      <td>Rinkimų programa LLRA-KŠS</td>\n",
       "      <td>Electoral program LLRA-KŠS</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>1</td>\n",
       "      <td>Lietuvos lenkų rinkimų akcijos – Krikščioniškų...</td>\n",
       "      <td>Lithuanian Polish Election Campaigns - Program...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>2</td>\n",
       "      <td>2016 m. Seimo rinkimai</td>\n",
       "      <td>2016 Seimas elections</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>3</td>\n",
       "      <td>Mieli Lietuvos žmonės!</td>\n",
       "      <td>Dear Lithuanian people!</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>4</td>\n",
       "      <td>Mūsų partijos veiklos pagrindas yra krikščioni...</td>\n",
       "      <td>The basis of our party's activities is Christi...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  author_affiliation author_name  para_no  \\\n",
       "0               LLRA        LLRA        0   \n",
       "1               LLRA        LLRA        1   \n",
       "2               LLRA        LLRA        2   \n",
       "3               LLRA        LLRA        3   \n",
       "4               LLRA        LLRA        4   \n",
       "\n",
       "                                           para_text  \\\n",
       "0                          Rinkimų programa LLRA-KŠS   \n",
       "1  Lietuvos lenkų rinkimų akcijos – Krikščioniškų...   \n",
       "2                             2016 m. Seimo rinkimai   \n",
       "3                             Mieli Lietuvos žmonės!   \n",
       "4  Mūsų partijos veiklos pagrindas yra krikščioni...   \n",
       "\n",
       "                                        en_para_text publish_time  \\\n",
       "0                         Electoral program LLRA-KŠS         2016   \n",
       "1  Lithuanian Polish Election Campaigns - Program...         2016   \n",
       "2                              2016 Seimas elections         2016   \n",
       "3                            Dear Lithuanian people!         2016   \n",
       "4  The basis of our party's activities is Christi...         2016   \n",
       "\n",
       "                     source  PC  AE  LogReg_ae  LogReg_pc  NaiveBaies_ae  \\\n",
       "0  party_election_manifesto   0   0          0          0              0   \n",
       "1  party_election_manifesto   0   0          0          0              0   \n",
       "2  party_election_manifesto   0   0          1          0              0   \n",
       "3  party_election_manifesto   1   0          0          1              0   \n",
       "4  party_election_manifesto   1   0          0          1              1   \n",
       "\n",
       "   NaiveBaies_pc  SVM_ae  SVM_pc  MLP_ae  MLP_pc  KNN_ae  KNN_pc  \n",
       "0              0       0       0       0       0       0       1  \n",
       "1              0       0       0       0       0       0       0  \n",
       "2              0       0       0       0       0       0       1  \n",
       "3              0       0       1       0       0       0       0  \n",
       "4              1       0       1       0       1       0       0  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction_df = pd.read_parquet(\"Data/PredictionDF.parquet\")\n",
    "print(prediction_df.shape)\n",
    "prediction_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13796, 5)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>publish_time</th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th>para_no</th>\n",
       "      <th>PC2</th>\n",
       "      <th>AE2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   publish_time author_affiliation  para_no  PC2  AE2\n",
       "0          2016               LLRA        0  0.0  0.0\n",
       "1          2016               LLRA        1  0.0  0.0\n",
       "2          2016               LLRA        2  0.0  0.0\n",
       "3          2016               LLRA        3  0.0  0.0\n",
       "4          2016               LLRA        4  1.0  0.0"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "true_data = pd.read_excel(\"Data/Hand_coded_LT.xlsx\")\n",
    "print(true_data.shape)\n",
    "true_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "true_data = true_data.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>publish_time</th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th>para_no</th>\n",
       "      <th>PC2</th>\n",
       "      <th>AE2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  publish_time author_affiliation  para_no  PC2  AE2\n",
       "0         2016               LLRA        0    0    0\n",
       "1         2016               LLRA        1    0    0\n",
       "2         2016               LLRA        2    0    0\n",
       "3         2016               LLRA        3    0    0\n",
       "4         2016               LLRA        4    1    0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "true_data = true_data.astype({\"publish_time\": str, \"PC2\": int, \"AE2\":int})\n",
    "true_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(13796, 21)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th>author_name</th>\n",
       "      <th>para_no</th>\n",
       "      <th>para_text</th>\n",
       "      <th>en_para_text</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>source</th>\n",
       "      <th>PC</th>\n",
       "      <th>AE</th>\n",
       "      <th>LogReg_ae</th>\n",
       "      <th>...</th>\n",
       "      <th>NaiveBaies_ae</th>\n",
       "      <th>NaiveBaies_pc</th>\n",
       "      <th>SVM_ae</th>\n",
       "      <th>SVM_pc</th>\n",
       "      <th>MLP_ae</th>\n",
       "      <th>MLP_pc</th>\n",
       "      <th>KNN_ae</th>\n",
       "      <th>KNN_pc</th>\n",
       "      <th>PC2</th>\n",
       "      <th>AE2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0</td>\n",
       "      <td>Rinkimų programa LLRA-KŠS</td>\n",
       "      <td>Electoral program LLRA-KŠS</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>1</td>\n",
       "      <td>Lietuvos lenkų rinkimų akcijos – Krikščioniškų...</td>\n",
       "      <td>Lithuanian Polish Election Campaigns - Program...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>2</td>\n",
       "      <td>2016 m. Seimo rinkimai</td>\n",
       "      <td>2016 Seimas elections</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>3</td>\n",
       "      <td>Mieli Lietuvos žmonės!</td>\n",
       "      <td>Dear Lithuanian people!</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LLRA</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>4</td>\n",
       "      <td>Mūsų partijos veiklos pagrindas yra krikščioni...</td>\n",
       "      <td>The basis of our party's activities is Christi...</td>\n",
       "      <td>2016</td>\n",
       "      <td>party_election_manifesto</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  author_affiliation author_name  para_no  \\\n",
       "0               LLRA        LLRA        0   \n",
       "1               LLRA        LLRA        1   \n",
       "2               LLRA        LLRA        2   \n",
       "3               LLRA        LLRA        3   \n",
       "4               LLRA        LLRA        4   \n",
       "\n",
       "                                           para_text  \\\n",
       "0                          Rinkimų programa LLRA-KŠS   \n",
       "1  Lietuvos lenkų rinkimų akcijos – Krikščioniškų...   \n",
       "2                             2016 m. Seimo rinkimai   \n",
       "3                             Mieli Lietuvos žmonės!   \n",
       "4  Mūsų partijos veiklos pagrindas yra krikščioni...   \n",
       "\n",
       "                                        en_para_text publish_time  \\\n",
       "0                         Electoral program LLRA-KŠS         2016   \n",
       "1  Lithuanian Polish Election Campaigns - Program...         2016   \n",
       "2                              2016 Seimas elections         2016   \n",
       "3                            Dear Lithuanian people!         2016   \n",
       "4  The basis of our party's activities is Christi...         2016   \n",
       "\n",
       "                     source  PC  AE  LogReg_ae  ...  NaiveBaies_ae  \\\n",
       "0  party_election_manifesto   0   0          0  ...              0   \n",
       "1  party_election_manifesto   0   0          0  ...              0   \n",
       "2  party_election_manifesto   0   0          1  ...              0   \n",
       "3  party_election_manifesto   1   0          0  ...              0   \n",
       "4  party_election_manifesto   1   0          0  ...              1   \n",
       "\n",
       "   NaiveBaies_pc  SVM_ae  SVM_pc  MLP_ae  MLP_pc  KNN_ae  KNN_pc  PC2  AE2  \n",
       "0              0       0       0       0       0       0       1    0    0  \n",
       "1              0       0       0       0       0       0       0    0    0  \n",
       "2              0       0       0       0       0       0       1    0    0  \n",
       "3              0       0       1       0       0       0       0    0    0  \n",
       "4              1       0       1       0       1       0       0    1    0  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.merge(left= prediction_df, right= true_data, on=[\"publish_time\", \"author_affiliation\", \"para_no\"], how=\"inner\")\n",
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds_ae = list(df[\"AE\"])\n",
    "preds_pc = list(df[\"PC\"])\n",
    "\n",
    "true_pc = list(df[\"PC2\"])\n",
    "true_ae=list(df[\"AE2\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6079012345679011\n",
      "0.4906666666666666\n"
     ]
    }
   ],
   "source": [
    "f1_pc = f1_score(true_pc, preds_pc)\n",
    "f1_ae = f1_score(true_ae, preds_ae)\n",
    "\n",
    "print(f1_pc)\n",
    "print(f1_ae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[10977   677]\n",
      " [  911  1231]]\n",
      "[[12947   298]\n",
      " [  275   276]]\n"
     ]
    }
   ],
   "source": [
    "conf_pc = confusion_matrix(true_pc, preds_pc)\n",
    "conf_ae = confusion_matrix(true_ae, preds_ae)\n",
    "\n",
    "print(conf_pc)\n",
    "print(conf_ae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.884894172223833\n",
      "0.9584662220933604\n"
     ]
    }
   ],
   "source": [
    "acc_pc = accuracy_score(true_pc, preds_pc)\n",
    "acc_ae = accuracy_score(true_ae, preds_ae)\n",
    "\n",
    "print(acc_pc)\n",
    "print(acc_ae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6451781970649895\n",
      "0.4808362369337979\n"
     ]
    }
   ],
   "source": [
    "prec_pc = precision_score(true_pc, preds_pc)\n",
    "prec_ae = precision_score(true_ae, preds_ae)\n",
    "\n",
    "print(prec_pc)\n",
    "print(prec_ae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5746965452847805\n",
      "0.5009074410163339\n"
     ]
    }
   ],
   "source": [
    "rec_pc = recall_score(true_pc, preds_pc)\n",
    "rec_ae = recall_score(true_ae, preds_ae)\n",
    "\n",
    "print(rec_pc)\n",
    "print(rec_ae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>party_year</th>\n",
       "      <th>f1_ae</th>\n",
       "      <th>f1_pc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2016_APKK</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2016_DK</td>\n",
       "      <td>0.625000</td>\n",
       "      <td>0.702703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2016_DP</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.627451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2016_LLP</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2016_LLRA</td>\n",
       "      <td>0.758621</td>\n",
       "      <td>0.656716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>2016_LLS</td>\n",
       "      <td>0.473684</td>\n",
       "      <td>0.656489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2016_LRLS</td>\n",
       "      <td>0.426230</td>\n",
       "      <td>0.603175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>2016_LS</td>\n",
       "      <td>0.642857</td>\n",
       "      <td>0.861111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2016_LSDP</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.566667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016_LVZS</td>\n",
       "      <td>0.461538</td>\n",
       "      <td>0.630769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2016_LZP</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2016_TAUT</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.857143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2016_TS-LKD</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.485261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016_TT</td>\n",
       "      <td>0.347826</td>\n",
       "      <td>0.514851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2020_DK</td>\n",
       "      <td>0.769231</td>\n",
       "      <td>0.754386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2020_DP</td>\n",
       "      <td>0.521739</td>\n",
       "      <td>0.634731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>2020_KS</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.676056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2020_KSS</td>\n",
       "      <td>0.222222</td>\n",
       "      <td>0.642857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2020_LLP</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.947368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020_LLRA</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>0.771084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2020_LP</td>\n",
       "      <td>0.297872</td>\n",
       "      <td>0.462312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2020_LRLS</td>\n",
       "      <td>0.392157</td>\n",
       "      <td>0.580475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2020_LS</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.763636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2020_LSDDP</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.578947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020_LSDP</td>\n",
       "      <td>0.342857</td>\n",
       "      <td>0.554839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>2020_LVZS</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.615385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020_LZP</td>\n",
       "      <td>0.285714</td>\n",
       "      <td>0.597015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2020_LaisTeis</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.710526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2020_LietuvaVisu</td>\n",
       "      <td>0.626263</td>\n",
       "      <td>0.748466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2020_NS</td>\n",
       "      <td>0.596491</td>\n",
       "      <td>0.675676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020_TS-LKD</td>\n",
       "      <td>0.217391</td>\n",
       "      <td>0.530466</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          party_year     f1_ae     f1_pc\n",
       "16         2016_APKK  0.800000  0.666667\n",
       "24           2016_DK  0.625000  0.702703\n",
       "17           2016_DP  0.166667  0.627451\n",
       "6           2016_LLP  0.200000  0.300000\n",
       "9          2016_LLRA  0.758621  0.656716\n",
       "29          2016_LLS  0.473684  0.656489\n",
       "26         2016_LRLS  0.426230  0.603175\n",
       "27           2016_LS  0.642857  0.861111\n",
       "8          2016_LSDP  0.500000  0.566667\n",
       "5          2016_LVZS  0.461538  0.630769\n",
       "10          2016_LZP  0.444444  0.529412\n",
       "18         2016_TAUT  0.500000  0.857143\n",
       "7        2016_TS-LKD  0.480000  0.485261\n",
       "0            2016_TT  0.347826  0.514851\n",
       "28           2020_DK  0.769231  0.754386\n",
       "11           2020_DP  0.521739  0.634731\n",
       "25           2020_KS  0.000000  0.676056\n",
       "14          2020_KSS  0.222222  0.642857\n",
       "19          2020_LLP  1.000000  0.947368\n",
       "2          2020_LLRA  0.533333  0.771084\n",
       "15           2020_LP  0.297872  0.462312\n",
       "23         2020_LRLS  0.392157  0.580475\n",
       "13           2020_LS  0.666667  0.763636\n",
       "22        2020_LSDDP  0.571429  0.578947\n",
       "3          2020_LSDP  0.342857  0.554839\n",
       "30         2020_LVZS  0.000000  0.615385\n",
       "1           2020_LZP  0.285714  0.597015\n",
       "20     2020_LaisTeis  0.250000  0.710526\n",
       "21  2020_LietuvaVisu  0.626263  0.748466\n",
       "12           2020_NS  0.596491  0.675676\n",
       "4        2020_TS-LKD  0.217391  0.530466"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"party_year\"] = df.apply(lambda x : str(x[\"publish_time\"])+\"_\"+str(x[\"author_affiliation\"]), axis = 1)\n",
    "party_years = list(set(df[\"party_year\"]))\n",
    "\n",
    "f1_by_party = []\n",
    "\n",
    "for py in party_years :\n",
    "    df_plh = df.loc[df[\"party_year\"] == py ]\n",
    "    true_ae = list(df_plh[\"AE2\"])\n",
    "    pred_ae = list(df_plh[\"AE\"])\n",
    "    f1_ae = f1_score(true_ae, pred_ae)\n",
    "    \n",
    "    true_pc = list(df_plh[\"PC2\"])\n",
    "    pred_pc = list(df_plh[\"PC\"])\n",
    "    f1_pc = f1_score(true_pc, pred_pc)\n",
    "    \n",
    "    d = {\"party_year\" : py , \"f1_ae\" : f1_ae, \"f1_pc\":f1_pc}\n",
    "    f1_by_party.append(d)\n",
    "\n",
    "df_f1 = pd.DataFrame(f1_by_party)\n",
    "df_f1= df_f1[[\"party_year\", \"f1_ae\", \"f1_pc\"]]\n",
    "df_f1=df_f1.sort_values([\"party_year\"])\n",
    "df_f1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-20-7fd45692bd22>:1: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n",
      "  df_year = df.groupby([\"publish_time\"])[\"AE\", \"AE2\", \"PC\", \"PC2\"].mean()\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AE</th>\n",
       "      <th>AE2</th>\n",
       "      <th>PC</th>\n",
       "      <th>PC2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>publish_time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016</th>\n",
       "      <td>0.039371</td>\n",
       "      <td>0.049251</td>\n",
       "      <td>0.131138</td>\n",
       "      <td>0.139521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020</th>\n",
       "      <td>0.043704</td>\n",
       "      <td>0.031197</td>\n",
       "      <td>0.145025</td>\n",
       "      <td>0.170039</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    AE       AE2        PC       PC2\n",
       "publish_time                                        \n",
       "2016          0.039371  0.049251  0.131138  0.139521\n",
       "2020          0.043704  0.031197  0.145025  0.170039"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_year = df.groupby([\"publish_time\"])[\"AE\", \"AE2\", \"PC\", \"PC2\"].mean()\n",
    "df_year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-21-5c2f2c5dbe75>:1: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n",
      "  df_agg = df.groupby([\"publish_time\", \"author_affiliation\"])[\"AE\", \"AE2\", \"PC\", \"PC2\"].mean()\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>AE</th>\n",
       "      <th>AE2</th>\n",
       "      <th>PC</th>\n",
       "      <th>PC2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>publish_time</th>\n",
       "      <th>author_affiliation</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"14\" valign=\"top\">2016</th>\n",
       "      <th>APKK</th>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DK</th>\n",
       "      <td>0.084507</td>\n",
       "      <td>0.140845</td>\n",
       "      <td>0.246479</td>\n",
       "      <td>0.274648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DP</th>\n",
       "      <td>0.008026</td>\n",
       "      <td>0.030498</td>\n",
       "      <td>0.102729</td>\n",
       "      <td>0.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LLP</th>\n",
       "      <td>0.011976</td>\n",
       "      <td>0.047904</td>\n",
       "      <td>0.053892</td>\n",
       "      <td>0.065868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LLRA</th>\n",
       "      <td>0.115044</td>\n",
       "      <td>0.141593</td>\n",
       "      <td>0.353982</td>\n",
       "      <td>0.238938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LLS</th>\n",
       "      <td>0.043478</td>\n",
       "      <td>0.053708</td>\n",
       "      <td>0.168798</td>\n",
       "      <td>0.166240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LRLS</th>\n",
       "      <td>0.029864</td>\n",
       "      <td>0.025339</td>\n",
       "      <td>0.103167</td>\n",
       "      <td>0.124887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LS</th>\n",
       "      <td>0.133333</td>\n",
       "      <td>0.177778</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LSDP</th>\n",
       "      <td>0.016878</td>\n",
       "      <td>0.008439</td>\n",
       "      <td>0.050633</td>\n",
       "      <td>0.075949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LVZS</th>\n",
       "      <td>0.040196</td>\n",
       "      <td>0.074510</td>\n",
       "      <td>0.160784</td>\n",
       "      <td>0.221569</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LZP</th>\n",
       "      <td>0.011236</td>\n",
       "      <td>0.039326</td>\n",
       "      <td>0.078652</td>\n",
       "      <td>0.112360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TAUT</th>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.444444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TS-LKD</th>\n",
       "      <td>0.056867</td>\n",
       "      <td>0.050429</td>\n",
       "      <td>0.137339</td>\n",
       "      <td>0.099249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TT</th>\n",
       "      <td>0.016260</td>\n",
       "      <td>0.030488</td>\n",
       "      <td>0.099593</td>\n",
       "      <td>0.105691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"17\" valign=\"top\">2020</th>\n",
       "      <th>DK</th>\n",
       "      <td>0.170616</td>\n",
       "      <td>0.199052</td>\n",
       "      <td>0.246445</td>\n",
       "      <td>0.293839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DP</th>\n",
       "      <td>0.016692</td>\n",
       "      <td>0.018209</td>\n",
       "      <td>0.109256</td>\n",
       "      <td>0.144158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KS</th>\n",
       "      <td>0.023810</td>\n",
       "      <td>0.011905</td>\n",
       "      <td>0.440476</td>\n",
       "      <td>0.404762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KSS</th>\n",
       "      <td>0.016340</td>\n",
       "      <td>0.042484</td>\n",
       "      <td>0.196078</td>\n",
       "      <td>0.261438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LLP</th>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.555556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LLRA</th>\n",
       "      <td>0.061728</td>\n",
       "      <td>0.030864</td>\n",
       "      <td>0.246914</td>\n",
       "      <td>0.265432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LP</th>\n",
       "      <td>0.034323</td>\n",
       "      <td>0.009276</td>\n",
       "      <td>0.095547</td>\n",
       "      <td>0.089054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LRLS</th>\n",
       "      <td>0.027226</td>\n",
       "      <td>0.010302</td>\n",
       "      <td>0.133922</td>\n",
       "      <td>0.144960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LS</th>\n",
       "      <td>0.116667</td>\n",
       "      <td>0.183333</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.450000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LSDDP</th>\n",
       "      <td>0.032558</td>\n",
       "      <td>0.016279</td>\n",
       "      <td>0.113953</td>\n",
       "      <td>0.151163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LSDP</th>\n",
       "      <td>0.078550</td>\n",
       "      <td>0.027190</td>\n",
       "      <td>0.208459</td>\n",
       "      <td>0.259819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LVZS</th>\n",
       "      <td>0.107143</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.178571</td>\n",
       "      <td>0.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LZP</th>\n",
       "      <td>0.011019</td>\n",
       "      <td>0.008264</td>\n",
       "      <td>0.063361</td>\n",
       "      <td>0.121212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LaisTeis</th>\n",
       "      <td>0.028369</td>\n",
       "      <td>0.028369</td>\n",
       "      <td>0.234043</td>\n",
       "      <td>0.304965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LietuvaVisu</th>\n",
       "      <td>0.244344</td>\n",
       "      <td>0.203620</td>\n",
       "      <td>0.361991</td>\n",
       "      <td>0.375566</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NS</th>\n",
       "      <td>0.084291</td>\n",
       "      <td>0.134100</td>\n",
       "      <td>0.249042</td>\n",
       "      <td>0.318008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TS-LKD</th>\n",
       "      <td>0.026353</td>\n",
       "      <td>0.006410</td>\n",
       "      <td>0.089031</td>\n",
       "      <td>0.109687</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       AE       AE2        PC       PC2\n",
       "publish_time author_affiliation                                        \n",
       "2016         APKK                0.250000  0.166667  0.166667  0.333333\n",
       "             DK                  0.084507  0.140845  0.246479  0.274648\n",
       "             DP                  0.008026  0.030498  0.102729  0.142857\n",
       "             LLP                 0.011976  0.047904  0.053892  0.065868\n",
       "             LLRA                0.115044  0.141593  0.353982  0.238938\n",
       "             LLS                 0.043478  0.053708  0.168798  0.166240\n",
       "             LRLS                0.029864  0.025339  0.103167  0.124887\n",
       "             LS                  0.133333  0.177778  0.400000  0.400000\n",
       "             LSDP                0.016878  0.008439  0.050633  0.075949\n",
       "             LVZS                0.040196  0.074510  0.160784  0.221569\n",
       "             LZP                 0.011236  0.039326  0.078652  0.112360\n",
       "             TAUT                0.111111  0.333333  0.333333  0.444444\n",
       "             TS-LKD              0.056867  0.050429  0.137339  0.099249\n",
       "             TT                  0.016260  0.030488  0.099593  0.105691\n",
       "2020         DK                  0.170616  0.199052  0.246445  0.293839\n",
       "             DP                  0.016692  0.018209  0.109256  0.144158\n",
       "             KS                  0.023810  0.011905  0.440476  0.404762\n",
       "             KSS                 0.016340  0.042484  0.196078  0.261438\n",
       "             LLP                 0.111111  0.111111  0.500000  0.555556\n",
       "             LLRA                0.061728  0.030864  0.246914  0.265432\n",
       "             LP                  0.034323  0.009276  0.095547  0.089054\n",
       "             LRLS                0.027226  0.010302  0.133922  0.144960\n",
       "             LS                  0.116667  0.183333  0.466667  0.450000\n",
       "             LSDDP               0.032558  0.016279  0.113953  0.151163\n",
       "             LSDP                0.078550  0.027190  0.208459  0.259819\n",
       "             LVZS                0.107143  0.000000  0.178571  0.285714\n",
       "             LZP                 0.011019  0.008264  0.063361  0.121212\n",
       "             LaisTeis            0.028369  0.028369  0.234043  0.304965\n",
       "             LietuvaVisu         0.244344  0.203620  0.361991  0.375566\n",
       "             NS                  0.084291  0.134100  0.249042  0.318008\n",
       "             TS-LKD              0.026353  0.006410  0.089031  0.109687"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_agg = df.groupby([\"publish_time\", \"author_affiliation\"])[\"AE\", \"AE2\", \"PC\", \"PC2\"].mean()\n",
    "df_agg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_agg2 = pd.DataFrame()\n",
    "df_agg2[\"year\"] = [i[0] for i in list(df_agg.index)]\n",
    "df_agg2[\"party\"] = [i[1] for i in list(df_agg.index)]\n",
    "df_agg2[\"AE\"] = list(df_agg[\"AE\"])\n",
    "df_agg2[\"AE2\"] = list(df_agg[\"AE2\"])\n",
    "df_agg2[\"PC\"] = list(df_agg[\"PC\"])\n",
    "df_agg2[\"PC2\"] = list(df_agg[\"PC2\"])\n",
    "df_agg2[\"paragraph_count\"] = list(df.groupby([\"publish_time\", \"author_affiliation\"])[\"para_no\"].max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>party</th>\n",
       "      <th>AE</th>\n",
       "      <th>AE2</th>\n",
       "      <th>PC</th>\n",
       "      <th>PC2</th>\n",
       "      <th>paragraph_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016</td>\n",
       "      <td>APKK</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016</td>\n",
       "      <td>DK</td>\n",
       "      <td>0.084507</td>\n",
       "      <td>0.140845</td>\n",
       "      <td>0.246479</td>\n",
       "      <td>0.274648</td>\n",
       "      <td>141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016</td>\n",
       "      <td>DP</td>\n",
       "      <td>0.008026</td>\n",
       "      <td>0.030498</td>\n",
       "      <td>0.102729</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLP</td>\n",
       "      <td>0.011976</td>\n",
       "      <td>0.047904</td>\n",
       "      <td>0.053892</td>\n",
       "      <td>0.065868</td>\n",
       "      <td>166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0.115044</td>\n",
       "      <td>0.141593</td>\n",
       "      <td>0.353982</td>\n",
       "      <td>0.238938</td>\n",
       "      <td>112</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year party        AE       AE2        PC       PC2  paragraph_count\n",
       "0  2016  APKK  0.250000  0.166667  0.166667  0.333333               11\n",
       "1  2016    DK  0.084507  0.140845  0.246479  0.274648              141\n",
       "2  2016    DP  0.008026  0.030498  0.102729  0.142857              622\n",
       "3  2016   LLP  0.011976  0.047904  0.053892  0.065868              166\n",
       "4  2016  LLRA  0.115044  0.141593  0.353982  0.238938              112"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_agg2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(31, 8)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>party</th>\n",
       "      <th>AE</th>\n",
       "      <th>AE2</th>\n",
       "      <th>PC</th>\n",
       "      <th>PC2</th>\n",
       "      <th>paragraph_count</th>\n",
       "      <th>party_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016</td>\n",
       "      <td>APKK</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>11</td>\n",
       "      <td>2016_APKK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016</td>\n",
       "      <td>DK</td>\n",
       "      <td>0.084507</td>\n",
       "      <td>0.140845</td>\n",
       "      <td>0.246479</td>\n",
       "      <td>0.274648</td>\n",
       "      <td>141</td>\n",
       "      <td>2016_DK</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016</td>\n",
       "      <td>DP</td>\n",
       "      <td>0.008026</td>\n",
       "      <td>0.030498</td>\n",
       "      <td>0.102729</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>622</td>\n",
       "      <td>2016_DP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLP</td>\n",
       "      <td>0.011976</td>\n",
       "      <td>0.047904</td>\n",
       "      <td>0.053892</td>\n",
       "      <td>0.065868</td>\n",
       "      <td>166</td>\n",
       "      <td>2016_LLP</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0.115044</td>\n",
       "      <td>0.141593</td>\n",
       "      <td>0.353982</td>\n",
       "      <td>0.238938</td>\n",
       "      <td>112</td>\n",
       "      <td>2016_LLRA</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year party        AE       AE2        PC       PC2  paragraph_count  \\\n",
       "0  2016  APKK  0.250000  0.166667  0.166667  0.333333               11   \n",
       "1  2016    DK  0.084507  0.140845  0.246479  0.274648              141   \n",
       "2  2016    DP  0.008026  0.030498  0.102729  0.142857              622   \n",
       "3  2016   LLP  0.011976  0.047904  0.053892  0.065868              166   \n",
       "4  2016  LLRA  0.115044  0.141593  0.353982  0.238938              112   \n",
       "\n",
       "  party_year  \n",
       "0  2016_APKK  \n",
       "1    2016_DK  \n",
       "2    2016_DP  \n",
       "3   2016_LLP  \n",
       "4  2016_LLRA  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_agg2[\"party_year\"] = df_agg2.apply(lambda x : str(x[\"year\"])+\"_\"+str(x[\"party\"]), axis=1)\n",
    "print(df_agg2.shape)\n",
    "df_agg2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(31, 10)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>party</th>\n",
       "      <th>AE Pred</th>\n",
       "      <th>AE True</th>\n",
       "      <th>PC Pred</th>\n",
       "      <th>PC True</th>\n",
       "      <th>paragraph_count</th>\n",
       "      <th>party_year</th>\n",
       "      <th>f1_ae</th>\n",
       "      <th>f1_pc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016</td>\n",
       "      <td>APKK</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>11</td>\n",
       "      <td>2016_APKK</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016</td>\n",
       "      <td>DK</td>\n",
       "      <td>0.084507</td>\n",
       "      <td>0.140845</td>\n",
       "      <td>0.246479</td>\n",
       "      <td>0.274648</td>\n",
       "      <td>141</td>\n",
       "      <td>2016_DK</td>\n",
       "      <td>0.625000</td>\n",
       "      <td>0.702703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016</td>\n",
       "      <td>DP</td>\n",
       "      <td>0.008026</td>\n",
       "      <td>0.030498</td>\n",
       "      <td>0.102729</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>622</td>\n",
       "      <td>2016_DP</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.627451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLP</td>\n",
       "      <td>0.011976</td>\n",
       "      <td>0.047904</td>\n",
       "      <td>0.053892</td>\n",
       "      <td>0.065868</td>\n",
       "      <td>166</td>\n",
       "      <td>2016_LLP</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0.115044</td>\n",
       "      <td>0.141593</td>\n",
       "      <td>0.353982</td>\n",
       "      <td>0.238938</td>\n",
       "      <td>112</td>\n",
       "      <td>2016_LLRA</td>\n",
       "      <td>0.758621</td>\n",
       "      <td>0.656716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016</td>\n",
       "      <td>LLS</td>\n",
       "      <td>0.043478</td>\n",
       "      <td>0.053708</td>\n",
       "      <td>0.168798</td>\n",
       "      <td>0.166240</td>\n",
       "      <td>390</td>\n",
       "      <td>2016_LLS</td>\n",
       "      <td>0.473684</td>\n",
       "      <td>0.656489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2016</td>\n",
       "      <td>LRLS</td>\n",
       "      <td>0.029864</td>\n",
       "      <td>0.025339</td>\n",
       "      <td>0.103167</td>\n",
       "      <td>0.124887</td>\n",
       "      <td>1104</td>\n",
       "      <td>2016_LRLS</td>\n",
       "      <td>0.426230</td>\n",
       "      <td>0.603175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2016</td>\n",
       "      <td>LS</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>0.177778</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>89</td>\n",
       "      <td>2016_LS</td>\n",
       "      <td>0.642857</td>\n",
       "      <td>0.861111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2016</td>\n",
       "      <td>LSDP</td>\n",
       "      <td>0.016878</td>\n",
       "      <td>0.008439</td>\n",
       "      <td>0.050633</td>\n",
       "      <td>0.075949</td>\n",
       "      <td>473</td>\n",
       "      <td>2016_LSDP</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.566667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2016</td>\n",
       "      <td>LVZS</td>\n",
       "      <td>0.040196</td>\n",
       "      <td>0.074510</td>\n",
       "      <td>0.160784</td>\n",
       "      <td>0.221569</td>\n",
       "      <td>1019</td>\n",
       "      <td>2016_LVZS</td>\n",
       "      <td>0.461538</td>\n",
       "      <td>0.630769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2016</td>\n",
       "      <td>LZP</td>\n",
       "      <td>0.011236</td>\n",
       "      <td>0.039326</td>\n",
       "      <td>0.078652</td>\n",
       "      <td>0.112360</td>\n",
       "      <td>177</td>\n",
       "      <td>2016_LZP</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.529412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2016</td>\n",
       "      <td>TAUT</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>8</td>\n",
       "      <td>2016_TAUT</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.857143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2016</td>\n",
       "      <td>TS-LKD</td>\n",
       "      <td>0.056867</td>\n",
       "      <td>0.050429</td>\n",
       "      <td>0.137339</td>\n",
       "      <td>0.099249</td>\n",
       "      <td>1863</td>\n",
       "      <td>2016_TS-LKD</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.485261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2016</td>\n",
       "      <td>TT</td>\n",
       "      <td>0.016260</td>\n",
       "      <td>0.030488</td>\n",
       "      <td>0.099593</td>\n",
       "      <td>0.105691</td>\n",
       "      <td>491</td>\n",
       "      <td>2016_TT</td>\n",
       "      <td>0.347826</td>\n",
       "      <td>0.514851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2020</td>\n",
       "      <td>DK</td>\n",
       "      <td>0.170616</td>\n",
       "      <td>0.199052</td>\n",
       "      <td>0.246445</td>\n",
       "      <td>0.293839</td>\n",
       "      <td>210</td>\n",
       "      <td>2020_DK</td>\n",
       "      <td>0.769231</td>\n",
       "      <td>0.754386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2020</td>\n",
       "      <td>DP</td>\n",
       "      <td>0.016692</td>\n",
       "      <td>0.018209</td>\n",
       "      <td>0.109256</td>\n",
       "      <td>0.144158</td>\n",
       "      <td>658</td>\n",
       "      <td>2020_DP</td>\n",
       "      <td>0.521739</td>\n",
       "      <td>0.634731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2020</td>\n",
       "      <td>KS</td>\n",
       "      <td>0.023810</td>\n",
       "      <td>0.011905</td>\n",
       "      <td>0.440476</td>\n",
       "      <td>0.404762</td>\n",
       "      <td>83</td>\n",
       "      <td>2020_KS</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.676056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2020</td>\n",
       "      <td>KSS</td>\n",
       "      <td>0.016340</td>\n",
       "      <td>0.042484</td>\n",
       "      <td>0.196078</td>\n",
       "      <td>0.261438</td>\n",
       "      <td>305</td>\n",
       "      <td>2020_KSS</td>\n",
       "      <td>0.222222</td>\n",
       "      <td>0.642857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2020</td>\n",
       "      <td>LLP</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.555556</td>\n",
       "      <td>17</td>\n",
       "      <td>2020_LLP</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.947368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2020</td>\n",
       "      <td>LLRA</td>\n",
       "      <td>0.061728</td>\n",
       "      <td>0.030864</td>\n",
       "      <td>0.246914</td>\n",
       "      <td>0.265432</td>\n",
       "      <td>163</td>\n",
       "      <td>2020_LLRA</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>0.771084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2020</td>\n",
       "      <td>LP</td>\n",
       "      <td>0.034323</td>\n",
       "      <td>0.009276</td>\n",
       "      <td>0.095547</td>\n",
       "      <td>0.089054</td>\n",
       "      <td>1078</td>\n",
       "      <td>2020_LP</td>\n",
       "      <td>0.297872</td>\n",
       "      <td>0.462312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2020</td>\n",
       "      <td>LRLS</td>\n",
       "      <td>0.027226</td>\n",
       "      <td>0.010302</td>\n",
       "      <td>0.133922</td>\n",
       "      <td>0.144960</td>\n",
       "      <td>1358</td>\n",
       "      <td>2020_LRLS</td>\n",
       "      <td>0.392157</td>\n",
       "      <td>0.580475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2020</td>\n",
       "      <td>LS</td>\n",
       "      <td>0.116667</td>\n",
       "      <td>0.183333</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.450000</td>\n",
       "      <td>59</td>\n",
       "      <td>2020_LS</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.763636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2020</td>\n",
       "      <td>LSDDP</td>\n",
       "      <td>0.032558</td>\n",
       "      <td>0.016279</td>\n",
       "      <td>0.113953</td>\n",
       "      <td>0.151163</td>\n",
       "      <td>429</td>\n",
       "      <td>2020_LSDDP</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.578947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2020</td>\n",
       "      <td>LSDP</td>\n",
       "      <td>0.078550</td>\n",
       "      <td>0.027190</td>\n",
       "      <td>0.208459</td>\n",
       "      <td>0.259819</td>\n",
       "      <td>330</td>\n",
       "      <td>2020_LSDP</td>\n",
       "      <td>0.342857</td>\n",
       "      <td>0.554839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>2020</td>\n",
       "      <td>LVZS</td>\n",
       "      <td>0.107143</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.178571</td>\n",
       "      <td>0.285714</td>\n",
       "      <td>27</td>\n",
       "      <td>2020_LVZS</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.615385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2020</td>\n",
       "      <td>LZP</td>\n",
       "      <td>0.011019</td>\n",
       "      <td>0.008264</td>\n",
       "      <td>0.063361</td>\n",
       "      <td>0.121212</td>\n",
       "      <td>362</td>\n",
       "      <td>2020_LZP</td>\n",
       "      <td>0.285714</td>\n",
       "      <td>0.597015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>2020</td>\n",
       "      <td>LaisTeis</td>\n",
       "      <td>0.028369</td>\n",
       "      <td>0.028369</td>\n",
       "      <td>0.234043</td>\n",
       "      <td>0.304965</td>\n",
       "      <td>140</td>\n",
       "      <td>2020_LaisTeis</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.710526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2020</td>\n",
       "      <td>LietuvaVisu</td>\n",
       "      <td>0.244344</td>\n",
       "      <td>0.203620</td>\n",
       "      <td>0.361991</td>\n",
       "      <td>0.375566</td>\n",
       "      <td>220</td>\n",
       "      <td>2020_LietuvaVisu</td>\n",
       "      <td>0.626263</td>\n",
       "      <td>0.748466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>2020</td>\n",
       "      <td>NS</td>\n",
       "      <td>0.084291</td>\n",
       "      <td>0.134100</td>\n",
       "      <td>0.249042</td>\n",
       "      <td>0.318008</td>\n",
       "      <td>260</td>\n",
       "      <td>2020_NS</td>\n",
       "      <td>0.596491</td>\n",
       "      <td>0.675676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>2020</td>\n",
       "      <td>TS-LKD</td>\n",
       "      <td>0.026353</td>\n",
       "      <td>0.006410</td>\n",
       "      <td>0.089031</td>\n",
       "      <td>0.109687</td>\n",
       "      <td>1403</td>\n",
       "      <td>2020_TS-LKD</td>\n",
       "      <td>0.217391</td>\n",
       "      <td>0.530466</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    year        party   AE Pred   AE True   PC Pred   PC True  \\\n",
       "0   2016         APKK  0.250000  0.166667  0.166667  0.333333   \n",
       "1   2016           DK  0.084507  0.140845  0.246479  0.274648   \n",
       "2   2016           DP  0.008026  0.030498  0.102729  0.142857   \n",
       "3   2016          LLP  0.011976  0.047904  0.053892  0.065868   \n",
       "4   2016         LLRA  0.115044  0.141593  0.353982  0.238938   \n",
       "5   2016          LLS  0.043478  0.053708  0.168798  0.166240   \n",
       "6   2016         LRLS  0.029864  0.025339  0.103167  0.124887   \n",
       "7   2016           LS  0.133333  0.177778  0.400000  0.400000   \n",
       "8   2016         LSDP  0.016878  0.008439  0.050633  0.075949   \n",
       "9   2016         LVZS  0.040196  0.074510  0.160784  0.221569   \n",
       "10  2016          LZP  0.011236  0.039326  0.078652  0.112360   \n",
       "11  2016         TAUT  0.111111  0.333333  0.333333  0.444444   \n",
       "12  2016       TS-LKD  0.056867  0.050429  0.137339  0.099249   \n",
       "13  2016           TT  0.016260  0.030488  0.099593  0.105691   \n",
       "14  2020           DK  0.170616  0.199052  0.246445  0.293839   \n",
       "15  2020           DP  0.016692  0.018209  0.109256  0.144158   \n",
       "16  2020           KS  0.023810  0.011905  0.440476  0.404762   \n",
       "17  2020          KSS  0.016340  0.042484  0.196078  0.261438   \n",
       "18  2020          LLP  0.111111  0.111111  0.500000  0.555556   \n",
       "19  2020         LLRA  0.061728  0.030864  0.246914  0.265432   \n",
       "20  2020           LP  0.034323  0.009276  0.095547  0.089054   \n",
       "21  2020         LRLS  0.027226  0.010302  0.133922  0.144960   \n",
       "22  2020           LS  0.116667  0.183333  0.466667  0.450000   \n",
       "23  2020        LSDDP  0.032558  0.016279  0.113953  0.151163   \n",
       "24  2020         LSDP  0.078550  0.027190  0.208459  0.259819   \n",
       "25  2020         LVZS  0.107143  0.000000  0.178571  0.285714   \n",
       "26  2020          LZP  0.011019  0.008264  0.063361  0.121212   \n",
       "27  2020     LaisTeis  0.028369  0.028369  0.234043  0.304965   \n",
       "28  2020  LietuvaVisu  0.244344  0.203620  0.361991  0.375566   \n",
       "29  2020           NS  0.084291  0.134100  0.249042  0.318008   \n",
       "30  2020       TS-LKD  0.026353  0.006410  0.089031  0.109687   \n",
       "\n",
       "    paragraph_count        party_year     f1_ae     f1_pc  \n",
       "0                11         2016_APKK  0.800000  0.666667  \n",
       "1               141           2016_DK  0.625000  0.702703  \n",
       "2               622           2016_DP  0.166667  0.627451  \n",
       "3               166          2016_LLP  0.200000  0.300000  \n",
       "4               112         2016_LLRA  0.758621  0.656716  \n",
       "5               390          2016_LLS  0.473684  0.656489  \n",
       "6              1104         2016_LRLS  0.426230  0.603175  \n",
       "7                89           2016_LS  0.642857  0.861111  \n",
       "8               473         2016_LSDP  0.500000  0.566667  \n",
       "9              1019         2016_LVZS  0.461538  0.630769  \n",
       "10              177          2016_LZP  0.444444  0.529412  \n",
       "11                8         2016_TAUT  0.500000  0.857143  \n",
       "12             1863       2016_TS-LKD  0.480000  0.485261  \n",
       "13              491           2016_TT  0.347826  0.514851  \n",
       "14              210           2020_DK  0.769231  0.754386  \n",
       "15              658           2020_DP  0.521739  0.634731  \n",
       "16               83           2020_KS  0.000000  0.676056  \n",
       "17              305          2020_KSS  0.222222  0.642857  \n",
       "18               17          2020_LLP  1.000000  0.947368  \n",
       "19              163         2020_LLRA  0.533333  0.771084  \n",
       "20             1078           2020_LP  0.297872  0.462312  \n",
       "21             1358         2020_LRLS  0.392157  0.580475  \n",
       "22               59           2020_LS  0.666667  0.763636  \n",
       "23              429        2020_LSDDP  0.571429  0.578947  \n",
       "24              330         2020_LSDP  0.342857  0.554839  \n",
       "25               27         2020_LVZS  0.000000  0.615385  \n",
       "26              362          2020_LZP  0.285714  0.597015  \n",
       "27              140     2020_LaisTeis  0.250000  0.710526  \n",
       "28              220  2020_LietuvaVisu  0.626263  0.748466  \n",
       "29              260           2020_NS  0.596491  0.675676  \n",
       "30             1403       2020_TS-LKD  0.217391  0.530466  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_agg3 = pd.merge(left= df_agg2, right=df_f1, on=\"party_year\", how=\"inner\")\n",
    "df_agg3 = df_agg3.rename(columns = {\"AE\" : \"AE Pred\", \"AE2\": \"AE True\", \"PC\" : \"PC Pred\", \"PC2\":\"PC True\"})\n",
    "print(df_agg3.shape)\n",
    "df_agg3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_agg3.to_pickle(\"Data/True_vs_pred.pkl\")\n",
    "df_agg3.to_excel(\"Data/True_vs_pred.xlsx\")\n",
    "df_agg3.to_excel(\"Annex/True_vs_pred.xlsx\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2016x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X = [i for i in list(df_agg2.index)]\n",
    "X2 = [i+0.25 for i in list(df_agg2.index)]\n",
    "\n",
    "# Declaring the figure or the plot (y, x) or (width, height)\n",
    "plt.figure(figsize=[28, 8])\n",
    "plt.xticks(rotation=45)\n",
    "plt.ylim(0, 0.6)\n",
    "\n",
    "plt.bar(X, list(df_agg2[\"AE\"]), color = 'crimson', width = 0.25)\n",
    "plt.bar(X2, list(df_agg2[\"AE2\"]), color = 'navy', width = 0.25)\n",
    "\n",
    "plt.legend(['Prediction', 'True Value', 'Active Cases'])\n",
    "# Overiding the x axis with the country names\n",
    "plt.xticks([i + 0.25 for i in range(len(df_agg2))], list(df_agg2[\"party_year\"]))\n",
    "# Giving the tilte for the plot\n",
    "plt.title(\"True and Predicted Values for AE\")\n",
    "# Namimg the x and y axis\n",
    "plt.xlabel('Party-Year')\n",
    "plt.ylabel('Share of Content')\n",
    "# Saving the plot as a 'png'\n",
    "plt.savefig('Figures/AE_TruePred.png')\n",
    "# Displaying the bar plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2016x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "X = [i for i in list(df_agg2.index)]\n",
    "X2 = [i+0.25 for i in list(df_agg2.index)]\n",
    "\n",
    "# Declaring the figure or the plot (y, x) or (width, height)\n",
    "plt.figure(figsize=[28, 8])\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "plt.bar(X, list(df_agg2[\"PC\"]), color = 'crimson', width = 0.25)\n",
    "plt.bar(X2, list(df_agg2[\"PC2\"]), color = 'navy', width = 0.25)\n",
    "\n",
    "plt.legend(['Prediction', 'True Value', 'Active Cases'])\n",
    "\n",
    "plt.xticks([i + 0.25 for i in range(len(df_agg2))], list(df_agg2[\"party_year\"]))\n",
    "# Giving the tilte for the plot\n",
    "plt.title(\"True and Predicted Values for PC\")\n",
    "# Namimg the x and y axis\n",
    "plt.xlabel('Party-Year')\n",
    "plt.ylabel('Share of Content')\n",
    "# Saving the plot as a 'png'\n",
    "plt.savefig('Figures/PC_TruePred.png')\n",
    "# Displaying the bar plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}
